The RESO Fall Conference in Austin, Texas was intensely busy this year, as the pace of both standards development and certification has accelerated.
Jeremy Crawford provided attendees with a crash course in the RESO organization. He described the various work groups, what each one does, and how to get involved and contribute. This is really important, because while some people like to complain about how slow RESO can be, RESO only goes as fast as volunteers make it go when they contribute to the work groups. So, people need to step up and get involved! Jeremy laid out a path for getting new ideas funneled through the Research & Development group, into an existing work group (e.g. Dictionary, Transport, RETS 1.x, PUID, Internet Tracking), through review of the Technical Committee, through the Board of Directors, to publication and compliance.
Greg Lemon provided attendees with an overview of the compliance process, including the next step after data dictionary compliance certification: Web API compliance certification. Two MLSs have already gotten their compliance to meet this NAR mandate, and I’ll write up more about the Web API in early 2016. Regarding data dictionary compliance, MLSs representing over 333,000 subscribers are compliant, and MLSs representing over 50,000 more are already in progress.
On day two, Jeremy Crawford and Mark Lesswing provided a “RESO State of the Standard” presentation. RESO 2015-2016 goals and themes were adoption and certification, leadership (increasing the membership and supporting the membership), and creation (dictionary, API, and PUID). In 2015 RESO delivered the Data Dictionary v1.4, a data dictionary wiki, a Data Dictionary certification program and MLS Data Dictionary certifications, the Web API v1.0.2, updated transport certification platforms, vendor transport certifications, and an organizational unique ID. In addition, RESO worked on industry leadership presence and outreach, and membership growth. Those are a lot of accomplishments!
RESO now has 227 members, up from 126 members in 2014. That’s 80% membership growth for the year to date! 58% of member organizations are MLSs, 31% are vendors, 9% brokerage, other associations are 1%, and NAR itself is 1%.
In 2016, the board of directors will include Mark Lesswing, Art Carter, Rob Overman, Tom Flanagan, Cary Sylvester, Rebecca Jensen, Chad Curry, Ethan Bailey, Justin LaJoie, Jeff Young, Brian Donnellan, Stuart White, Craig Cheatham, and Mark Wise. There is an open MLS seat and an open vendor seat, and there will soon be an election for those seats.
Current workgroups (and chair-people) are as follows:
- Research and Development (Greg Moore)
- Data Dictionary (Rob Larson)
- Internet Tracking (Chris Lambrou)
- Transport (Scott Petronis)
- RETS 1.x (Paul Stusiak)
- Property Unique ID (Paul Desormeaux)
There are also three committees (and committee chair-people):
- Technical Committee (Chad Curry)
- Certification Committee (Greg Lemon)
- Marketing Committee (Jeremy Crawford)
A panel, moderated by Mark Lesswing, discussed the RESO 2015-2016 strategic vision and what is next for the standards and RESO. Jeremy Crawford mentioned “Update” via the new Web API – making the standard more bi-directional rather than just being a way to send data out of the MLS – and a continued focus on enhancing the Data Dictionary so that every field is referred to the same way (technically) all over the country. Michael Wurzer talked about the need for localization and the need to represent localizations in a standard way. Rebecca Jensen talked about getting brokers involved to help drive discussion of standards-related business rules forward. Jeremy mentioned that regionally, some MLSs are conforming their business rules, especially their data oriented rules. Rebecca Jensen talked about the challenge of developing and getting adoption for a standard on a budget. Michael expressed the need for a unique property ID to tie together all the disparate databases that we have that surround properties. The group then discussed about how RESO has to work to meet broker needs, and how some of RESO’s evolving standards will help do that, helping MLSs continue to facilitate participant cooperation.
Additional Business Presentations and “Pre-check”
I couldn’t attend the rest of the business track because of the concurrent technical sessions, but I heard great things about some of the sessions, including broker discussions and demonstrations of how they are using RETS, discussions about adopting the RESO data dictionary, and data licensing best practices. But, one of the cool ideas that came out of the business sessions was the idea of a “Vendor Pre-check”. One person later suggested that I could review a vendor’s product with relation to typical MLS policy and also look at security practices, someone else could review their business practices, and a few MLSs could provide their additional input to certify a vendor “Pre-checked.” But maybe it could be simpler, where, once a certain number of MLSs indicated that the vendor was in current good standing with them (say, with CMLS?) other MLSs would know that diligence had probably been performed, the vendor would have “Pre-checked” status and they would have an easier time getting approved for data access at other MLSs. This probably requires more thought – for example, how to deal with negative reviews by MLSs fairly – but I like the idea, in general.
Great Presentation by CRT
Chad Curry, Chris Coté and Dave Conroy from CRT presented about the Internet of Things in real estate and their new “CRTLabs” (http://crtlabs.org) project, which will enable CRT to experiment with “things” rather than just with software and technical practices. The Internet of Things will enable NAR members to improve the quality of life for everyone. Using sensors for CO2, temperature, energy, and light, people can monitor and create better living environments. This type of dynamic data could supplement the static data that is input into the MLS. Perhaps as a closing gift agents could provide an environmental sensor system and other smart items that improve the buyer’s quality of life. Such gifts could help redefine the relationship of agents with buyers, outside of the transaction. CRT is partnering with universities, NGOs, and vendors on its Internet of Things initiative. Ideally some of these items could become member benefits, and members could act as a focus group around some of these products. CRT is also working on security initiatives to help both vendors and homeowners manage security around the Internet of Things.
Chris Cote described OpenADR (automated device response) – a system under development that will help utilities maintain grid reliability and help customers control their energy future. He also gave us a peek at “Rosetta Home” – open source software that CRT is developing to gather data from various sensors and give a consumer a dashboard view into all the different kinds of data that can be created about their home environment. This was also shown during the RESO PlugFest event. Finally, Chris described the “Array of Things” – city wide networks to measure air quality, traffic counting, heat, standing water, and so much more. I wrote about some of the applications and opportunities for this in my blog written after the CMLS conference earlier this year. Dave Conroy demonstrated “Rosetta Home,” forcefully showing how the Internet of Things data could be integrated into an MLS listing report. Hopefully, members can become a resource on home environment and monitoring for homeowners and homebuyers, and thereby help enhance their communities.
I want to give a special shout-out to Chris and Dave – both of whom are fairly new CRT additions – great job, guys!
Department of Energy
The Information Exchange Models Working Group met to track concepts and clarify common definitions, data sets, and transfer methods for applying home energy information to real estate use cases. The group discussed how to overlay the Building Energy Data Exchange Specification (BEDES, pronounced “beads”) overlay on top of RESO’s standards. The Data Access and Security Working Group met to identify barriers and seek joint solutions related to permissions and legal issues in sharing home energy information and best practices in keeping data secure. There are three approaches to leveraging BEDES, and RESO can evaluate using one or more of the following: including BEDES energy related fields in the RESO Data Dictionary, mapping our terms and BEDES’s terms, and/or using BEDES terms as “synonyms” in our data exchange formats. The group will continue to publish new mappings to serve as a “Rosetta stone” for BEDES and our standards to encourage adoption, and it will continue to support mashups of the data. For more information on BEDES: http://energy.gov/eere/buildings/bedes-current-activities
New Standards Technology Initiative in Austin, TX!
Tim Dain (new MLS Director of ABoR) and Ethan Bailey (CoreLogic) took the stage to announce a strategic technology initiative. They are building something that they can deliver to real estate startups to encourage innovation – a fully functional reference data server that provides real-time data and transactional information. It will be free to those startups, and based on CoreLogic’s Trestle™ Platform. The platform is currently RETS 1.8 transport compliant and RESO data dictionary 1.4 bronze level compliant. This should be delivered in Q1 2016. Of course, in 2016 the solution will use the new RESO Web API and the platform will evolve as the standard evolves. There will be more information available on http://reso.org as it becomes available.
The PlugFest was an opportunity for developers to create and present concepts of new ways that RETS data could be used in combination with new technologies and other data sources. Rob Overman, CTO at Lone Wolf, emceed this session, and the session was sponsored by the Austin Board of Realtors. There were many great presentations – frankly, you had to be there to get the most out of this session. A group called ImageIndexr demonstrated some new interfaces for viewing listings based on a more image-oriented approach – kind of “Pinterest meets Amazon.com”. In such an interface, users might rate their favorite kitchens from a group of photos from various homes meeting search criteria, then rate the living rooms, then bathrooms, etc. – and then the system would know which homes they found most appealing overall, and could also present them with a “people who liked these photos also liked these houses”. This is an interface I’d seen before in various contexts, but their implementation was appealing. Note that ImageIndexr also talked about tagging images to more easily find misused images online, but as industry insiders already know most stolen content isn’t put back online to find, and in the rare occasion that it is, the reactive and expensive legal battle can be ugly even once ownership of image copyright is established. ImageIndexr won third place in the competition. I especially enjoyed seeing how CRT combined home energy and environmental sensors with listing data to show how both home-searching consumers and homeowners might be presented with this kind of content in an appealing manner. CRTLabs won second place in the competition. The first place winner of the PlugFest was Team Particle, which was an ad-hoc team that came together at the event – unfortunately they did not have time provided to show their work to the larger group meeting.
R&D Work Group
This work group examines the business cases for RESO taking on new work and, at least to some extent, vets the technical aspects of those potential projects. As I discussed earlier, R&D is the “funnel” for getting big new ideas into RESO standards. Following are some of the items we discussed:
Multifamily Information and Transaction Standards (MITS Standards). The MITS standards mostly relate to apartment and multi-family properties, especially property management. It could be an interesting supplement to RETS efforts. RESO has been asked if it is interested in bringing this set of standards under the RESO organization. The work group believes this is a good idea and has suggested that additional legal and other diligence be performed. For more about the MITS standards, see http://mitsproject.org/
Organizational Unique ID (OUID). This effort is about having a consistent way to refer to organizations in our industry, especially data sources such as MLSs, vendors, boards, and associations. It could be extended to franchisors, brokers, content providers, appointment systems, lockbox companies and more. It could be a valuable part of helping listing aggregators de-duping or stacking listings. From a systems integrator perspective, having multiple parties referring to MLSs and other parties the same way is also useful. Such an identifier could also be used to manage the chain of listing provenance documentation – to know the true source of data and steps between them and an aggregator/portal, especially if the aggregator/portal gets it from several sources. RESO has collected vendor/MLS organizational information in a spreadsheet and is working on creating a unique ID for each. The group discussed how to make it more manageable and how to best expose this data in a more programmatically useful format. Perhaps a push notification of changes? Or a way to pull it via RETS? The primary action item recommended is that the OUID become a part of the data dictionary and certification. Second, it must be made manageable online. Third, RESO must create a better way to consume the information than the current spreadsheet.
Media. Creating additional information around the images transported by RESO standards could enable consumers of those images to use them more easily and effectively. There is some interest in creating standards for tagging images by size (e.g. thumbnail, medium, large, best…), as well as a more standardized way of describing what part of the property is being portrayed in an image. Other possible fields include orientation, aspect ratio, white padded, cropped, branded and more. Most of these ideas have not been prioritized by the workgroup, but there is definitely a desire to send information around copyright and licensing along with the image. That initiative may be taken further by the data dictionary workgroup, considered in consultation with RESO’s attorney, Mitch Skinner. In the future we might look at how to address video and stitched image formats (walkthroughs, etc.).
Using One Login for Multiple Feeds. There’s a need to promote a best practice to expose active user-ids and statuses to vendors so they can ensure they don’t misuse data across their customers, when for efficiency’s sake they download data once for multiple MLS-approved customers and uses. Since there seems to be sensitivity at some MLSs to providing the roster resources to some vendors, potentially the group could recommend creating a subset of the roster resource that is more limited than the roster – just agent ID / name, office identifier, and status. I suggested that we could create a system for vendors to use a ‘vendor account’ associated with non-interactive accounts with their associated privileges (as unpacked better in my 2014 blog that initiated this whole topic) – but this adds complexity. There’s also a discussion of providing a way for someone pulling RETS data to know what’s usable for different usages (IDX, VOW, back office, etc.), which is a bit of a different issue that requires more discussion.
At the work group meeting I brought up the need to re-engage on the subject of creating a means for RESO standards to document and transmit MLS data business rules, so that vendors can easily obtain all of the more complex business rules from an MLS to facilitate data updates and system transitions. This is something I documented thoroughly for the group in 2010 and have brought up a few times since. The need for it continues to grow and become more urgent. The new work group chair, Greg Moore, understands the need to get this going, so I have some hope that RESO will move forward on this initiative.
Property Unique Identifier (PUID) Work Group
Having a Property Unique Identifier (PUID) for each property is useful for all sorts of reasons – de-duplication (for MLSs vendors, websites, etc.), aggregation (tying to mash up multiple sources), fraud prevention (uniqueness of mortgage contracts), property history reports / CDOM (even if the address has a “typo” in it), and Internet Tracking (providing value to MLSs, vendors, brokers, agents…). The original thinking was that, to solve this problem, we could simply “slap a few fields together” to generate an identifier. It’s still a possibility. But, that might only work for 95% of the cases. There are lots of cases where we don’t all have the data at hand reliably, and even APNs (tax IDs) are sometimes re-used or never assigned to begin with. Could we step back and use other pieces of information – things that MLSs have in their database? For example: formula fields (tax fields, unit number, etc.), geography (i.e. lat, long, elevation, entry level, etc.) and address information (city, country, county or parish, postal, street related fields, lot size, etc.)? Rather than generating a simple unique ID – or in addition to it – we could create one field where we concatenate a bunch of other fields and we can create a “compare PUID” function, where even if a property is missing one part of the field or has some parts that are not quite the same, we could determine how likely it is that the properties are one and the same. We need to consider what the algorithm for this might be; then we can create pseudo-code and potentially a reference implementation. Perhaps there could be a web API that calls this implementation, although there is a concern that this latter approach is too “Rube Goldberg.” Theoretically we could also then generate a smaller PUID that refers to this concatenated field in some way. Mark Lesswing also suggested we look at the financial industry’s use of block chaining, and employ it as a PUID mechanism; this approach is used in Honduras and could be used not only to identify a property but also to track its history.
Internet Tracking Work Group
This is a brand new work group that is focused on creating standards for moving data related to online activity – for example, how many people have viewed a listing, on what sites, how they got to that listing, and how they interacted with that listing. Once standards are created, it will be far easier to compile a better and comprehensive picture of what is happening around listings online. This group’s main activity during this meeting was to start to lay out some of the business cases for the different kinds of information that should be standardized. Chris Lambrou, the chair of this group, made a presentation to the technical track attendees to describe what the group is doing. The group has identified stakeholders: who will take this data and how will they use it (i.e., what are the business use cases)? This drives what fields are needed to support those use cases. The group will also look at current solutions, from Zillow to SourceMLS, to find a standard that will work for everyone.
Transport Work Group
This group met at the same time as the PUID group, and Jeremy Crawford told me that the main activity for the group during this meeting would be to discuss compliance requirements. No doubt the output of this meeting will be documented for wider release, as we are quickly moving toward a time when MLSs and their vendors will need to know a lot more about transport compliance.
A panel presented its experience with RETS 1x data dictionary compliance. The question was whether to do a native mapping, rename existing fields to the data dictionary names, or map through a proxy to avoid making local changes. Rob Larson said that he’s using both: a proxy that “gets him in the door” – that allows people to use the new while he transitions the current database and forms to the new dictionary terms, and while he transitions RETS users to native dictionary use. Rick Trevino is changing what’s happening in the background to use the dictionary, but is not changing input forms or reports. Richard Renton is using CoreLogic Trestle as a proxy, which gave him instant certification, but he’s also working with Bridge Interactive as well. Richard brought up some of the complexity that can attend compliance – for example, his MLS doesn’t – by rule – publish square footage, which is a core field in the data dictionary. So, compliance can be more than a technical endeavor. Nonetheless, Rob made a great point: that if you don’t have data for a field – even a “core” field – you don’t have to add the field to be compliant. Paul Stusiak presented further about RETS 1.x to the larger meeting, explaining the history of 1.x all the way to the current version, 1.8.1, along with the changes made in each version. In the future, there is an expectation that there will be further changes – one relates to IETF (Internet Engineering Task Force) changes, and the other is changing from DTD to XSD. Otherwise, community driven change is made through the RETS Change Proposal process.
Day Three Technical Discussions
Data Dictionary Version 1.5 is in progress. There are 234 fields and many changes including some field length and certification level changes; also eight field additions and other minor changes. There are now 45 enumerations, including 18 additions. The plan is to enumerate 55 more fields before the spring conference. The new version will also include standards for saved searches (leveraging OData’s $filter capability), lists (i.e., watched listings, favorites), ownership enumeration (i.e. condo, fee simple), energy (the more useful BEDES fields, adding a way to pass-through technical data), and accessibility. There is also the definition of an IDX payload and it was discussed whether a number of fields should be required for a dictionary or separate IDX certification; MLSs should provide feedback on the proposed fields in the RESO discussion forums. There’s a new Data Dictionary Wiki under development (not yet released) which will help people get acclimated to terms and fields relating to the transmission of real estate data. There will be an in-person meeting to work on enumerations in San Dimas, CA – December 8-9.
Builder’s Data Overview. Tim Costello, Chief Executive Officer of BDX, a builders’ consortium, provided an update about the data standards they have around the new home industry and how this content may fit within the RESO Data Dictionary. While 56% of all home shoppers consider new construction and 20% enter the process believing they are going to buy a new home, agents often don’t have expertise in new homes – or access to the new home content in their systems. Also, currently this content doesn’t fit well into the traditional “existing home” search taxonomy and new home data remains “highly compromised” in nearly all existing search sites. New home searches focus on finding a community (NOT an address search), discovering a plan and elevation, selecting a lot (requiring an interactive plat map), configuring the home to meet their needs (it could be a bedroom, or made into another room!) and personalizing it – that’s very different from a traditional existing home search where, for example, the number of bedrooms and bathrooms is already set. Because of all these options, the price is also fluid and there are also dynamic retail incentives and promotions, again, unlike existing home sales. All of this drives different data needs and search process. Even if we could harmonize this data with RESO data dictionary there would still be a lot of unique data needed to support new construction, especially since 80% of the content is plan based – not based on pre-built new construction. Going forward, BDX wants to dialogue with RESO, synchronize data standards and feeds, and assist in the deployment of new home search platforms, architected around community-based, non-deterministic search. Should our existing platforms be enhanced to provide for better new home search, such that we would need to work on moving this additional data around, then integrate existing platforms with a specialized system that allows for the activities of customization etc.? At the end of the session, Jeremy reminded the room that, “We’re not the Old Home Standards Organization, we’re the Real Estate Standards Organization.”
Making a Global Impact with the RESO Data Dictionary. Matt Kumar and Tony Satyadas from Innovation Incubator talked about what they are doing with the RESO Data Dictionary in the rapidly developing real estate industry in India. They have a need for this standardization, though there are some unique factors in India – additional fields needed for “SmartCities”, unique forms of consumer interactions, and changes made to reflect the Indian real estate vocabulary and business practices. They are also looking to innovate – adding data to support drones, Internet of Things, and more; also to create an iRESO certification program for the Indian RE ecosystem.
RESO Web API Status Update. Scott Petronis from Onboard Informatics provided this update. This API will provide a common, modern method for developers to access data. It’s focused on real-time access, replication, adds, changes, and deletes – all using the same technology model. It should reduce development time, complexity, and cost, and for faster prototyping, iterating, and time to market. The standard has been published and approved by the board, currently at version 1.0.2. It’s based on OData 4.0, an OASIS standard. Currently it’s focused on read access and search. The current security model is OAuth 2.0, which is what the certification is based on currently. However, this will most likely be changed to OpenID Connect, which is built on top of OAuth2. The group is still finalizing the Entity Data Model that facilitates data dictionary usage. Adoption is mandated by NAR for June 30, 2016. There will be certification documentation by the end of the year and certification testing will begin in Q1 2016. The group needs volunteers to help work on the next major API initiative – focused on “Update” (write) capabilities. The targets for this is to establish use cases, propose an approach by end of year, draft a standard by the end of Q1, and target approval by mid-year.
RESO Web API Security. Cal Heldenbrand from FBS, and Ashish Antal and Maria Dalarcao from MLSListings, demonstrated how OpenID connect can be used to provide authentication for the new RESO Web API. OAuth2 based flow allows for user authorization, security tokens can transfer claims and identity to web services, and the JSON Web Tokens (JWT) model reduces dependencies on back-end for claims verification.
In Conclusion / Other Resources
There are some great resources for MLSs and brokers to help them understand why RESO, participation, and certification are important.
There were several tracks, so I couldn’t cover absolutely everything that happened at the event, but hopefully this post helps you get a sense for how much value there is at RESO and its events. This was a busy conference and it was great to see such great participation this year from over 230 attendees. If you’re not already a RESO member, I would strongly recommend becoming one and attending the conference!
Share this post: